Conversation
JTaky
changed the title
There is no continuous per-(table, subtask) view of write throughput: CommitMetrics counters only update when a commit lands, so they go dark during checkpoint stalls that need diagnosing, and every writerBuffer gauge except numWriters reports the shared per-subtask buffer pool rather than anything per-table. Register a rowsWritten counter on the writer operator's MetricGroup under a per-table metric group, so it is tagged with both table and subtask and is polled independently of checkpointing. Counted on the returned record so rows dropped by the row-kind filter (ignore-delete) are excluded, and in all three write() overloads since they do not delegate to each other. Rows are counted entering the write buffer, making this an upper bound on
JTaky
force-pushed
the
oleksandr.nitavskyi/writer-row-metrics
branch
from
September 16, 2026 11:43
01de283 to
d28f654
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Purpose
There is no continuous per-(table, subtask) view of write throughput: CommitMetrics counters only update when a commit lands, so they go dark during checkpoint stalls that need diagnosing, and every writerBuffer gauge except numWriters reports the shared per-subtask buffer pool rather than anything per-table.
Register a rowsWritten counter on the writer operator's MetricGroup under a per-table metric group, so it is tagged with both table and subtask and is polled independently of checkpointing.
Tests
Deployed run in prod for us and caught some tables with high churn which would benefit from more buckets